Ayar Labs

Emma Steuer 🧚🤖

@emmysteuer

The thing slowing down AI isn't necessarily the chips. It's the copper wires running between them Above a certain bandwidth, copper loses data, generates heat, and runs out of physical room to scale. Ayar Labs just raised $500M, with NVIDIA and AMD among the investors, to replace those copper interconnects with light 8 terabits per second at up to 20x more performance per watt. Same principle as fiber optic cables, just running between processors inside the same package instead of between buildings

Serenity

@aleabitoreddit

It's slightly more nuanced. Basically, $AMD SF event is tomorrow, Wed/Thurs (July 22, 23). From Jefferies client note, AMD is expected to provide additional disclosures around their likely scale-up CPO roadmap + MI500. If they confirm CPO route, would positive thematically since it's not just $NVDA forcing the transition. As for $SIVE, we'll get a better read if they're directly in AMD supply chains. Since right now, it's possible to map it through $GFS (reportedly involved in AMD MI500 CPO work) or AYAR (which AMD invested in). But no formal confirmations. So if you're turning in: - CPO thematic positive: MI500 explicitly commits to CPO/optical UALink. - AMD gives details using Ayar Labs or GF SCALE for MI500 CPO: That would be extremely positive for $SIVE, given they're the upstream laser supplier / reference laser. Currently, Blayne Curtis from Jefferies expect AMD to work with $ALAB on UAL and $AVGO on ESUN... but juicy details are here: "We are looking for whether AMD commits to a CPO based scale up solution and who supplies the optical engine" We'll get a broader read through on the supply chain come Wednesday/Thursday, too early to tell today.

Mike

@MikeLongTerm

BREAKING $AMD| Jefferies analyst Blayne Curtis (top 1% ranked) is very Bullish @AMD 's upcoming "Advancing AI 2026" event on July 22–23 in San Francisco Fun fact: Blayne Curtis is still delaying his PT update, we will probably see his comment/note after Advancing AI event. Jefferies' bullish Expectations: ~CPU TAM Expansion: AMD could raise its addressable market above Nvidia's prior $200B figure, driven by AI workloads ( agentic AI boosting server CPU demand). ~MI500 GPU Details: More on the CDNA 6-based family (previewed at CES 2026: 2nm process, HBM4E memory, 2027 launch, claimed 1,000x AI performance vs. MI300X 8-GPU node). Focus on native UAL scale-up (up to 256 GPUs/rack) and potential optical interconnects (CPO/co-packaged optics). ~Networking/Optical: Partnerships with Astera Labs (UAL), Broadcom (ESUN switches), and possible Ayar Labs involvement. Optical confirmation would be positive for the supply chain. ~Other Bullish catalysts: Customer announcements. Existing wins include OpenAI and Meta (both ~6GW deals with incentives). Jefferies sees Microsoft as a MI450 customer (per supply chain checks) and potential for Anthropic (hiring ROCm engineers signals diversification). A "traditional" (less incentivized) deal with Anthropic would be very positive

Anders Storm

@StormDirac

"Is Optical Scale-Up Finally Approaching? NVLink Fusion hints that optics may be moving inside the rack" by Knvul Sheikh. Sheikh is a science journalist who covers on AI, biotechnology and the semiconductor industry. "Ayar Labs, one of the photonics companies participating in the NVLink Fusion ecosystem, has developed optical chiplets meant to sit alongside GPUs and other processors, converting electrical signals into light only millimeters from the compute silicon. “The most optimal way to do that is having a photonic chiplet with an electronic chiplet and having them hybrid bonded together,” says Ayar’s director of product management, Vishal Chandrasekar. He argues that optical scale-up has become practical because the semiconductor manufacturing ecosystem has matured around co-packaged optics. Advances in hybrid bonding now allow electronic and photonic chiplets to be manufactured separately, then integrated into a single optical engine that sits beside GPUs or switches. “What has really happened in the last few years is that the process maturity coming out of the fabs has really improved,” Chandrasekar says. He believes the industry is now on a path toward high-volume optical scale-up systems within the next couple of years. “Absolutely,” Chandrasekar says. “There are going to be multiple implementations in the 2028 time frame in very high volume.” https://t.co/k1VrybqDUn

revoAIution

@revoAIution

Morgan Stanley's scale-up market research says recent selloffs surrounding delayed CPO adoption are viewed as overdone. Relevant for $SIVE holders even though this report predates the current drawdown. Worth being straight about the timeline too. Morgan Stanley sees meaningful CPO adoption really starting with Nvidia's Feynman generation, 2028 to 2029, later than the more optimistic 2027 windows tied to some of $SIVE's specific connections. Copper is holding on longer in scale up than some bulls have been pricing in. But the addressable market keeps growing regardless of exact timing. Scale up network market projected at $70 billion plus by 2030, four times larger than the estimate from just a year ago. And NVLink Fusion, the same standard Ayar Labs joined in June with $SIVE as their confirmed laser supplier, is described here as the mechanism letting third party chips tie directly into Nvidia's ecosystem.

Athu

@athuinvests

Optical engines are a key CPO constraint, according to TrendForce, largely due to the shortage of InP lasers. Especially since $NVDA locked in capacity with $COHR and $LITE. > Ayar Labs builds optical engines powered by $SIVE InP CW lasers. > Expecting POs in 2027-28 for $POET CPO solutions that also use $SIVE lasers, or $GFS, which is partnering with $AMD for its own CPO program. > $JBL is more for 1.6T+ programs, not CPO. The CEO mentioned them trying to gain more market share and focusing on pluggables in the short term, so $SIVE could get more customers too. $SIVE needs to (IMPORTANT): - qualify their lasers for production from the development stage - improve their laser technology to gain a lead - ensure enough capacity to fulfill demand As their CEO mentions: In an industry where you’re supply constrained and there is huge demand, competition becomes a smaller problem.

Alex Wright-Gladstein

Co-founder

Co-founded Ayar Labs and served as its first CEO.